pypomp.PanelPomp.simulate

PanelPomp.simulate(key: Array, theta: PanelParameters | None = None, times: Array | None = None, nsim: int = 1, as_pomp: bool = False) → tuple[DataFrame, DataFrame] | object

Simulate latent states and observations from the panel model.

Parameters:

  • key (jax.Array) – JAX random key.

  • theta (PanelParameters or None, optional) – Parameters to simulate from. If None, defaults to self.theta.

  • times (jax.Array or None, optional) – Times at which to simulate the model. If None, defaults to the times coordinate of the data.

  • nsim (int, optional) – Number of simulations to run per replicate. Defaults to 1.

  • as_pomp (bool, optional) – If True, return a new PanelPomp object containing the simulated observations for the first parameter replicate and simulation index. Defaults to False.

Returns:

If as_pomp=False, returns a tuple (X_sims, Y_sims) of dataframes in long format. If as_pomp=True, returns a deep copy of the original panel model with simulated observations.

Return type:

tuple of (pd.DataFrame, pd.DataFrame) or PanelPomp